House Sales Prices and Venues in Municipal Parts of Prague

A.Introduction/Business Problem

Prague is one of the most popular city in Europe with many outstanding views and historic, exciting structures. My journey of Prague has started with the hiring of my wife by a global company there. In this study, I give a brief insight about the population, house prices and types, venues and compare and cluster different municipal parts of the city for anyone looking for buying a property in Prague or interested in real estate industry.

B.Data Prepareation & Description

The data is used in this project can be divided into three categories.

-Population by Municipal Parts: I took the table from ... as csv.

-House sales price: I scrape all available advertisements for Prague in SReality.com which is one the most used webpage for property sale. (around 4700 objects)

-Venues data in municipal parts: I used Foursquare API for getting the data.

B1.Population by Municipal Parts

Here is the population data for 57 Municipal Parts of Prague.

B2.House sales price

Sreality.com is one of the most used advirtisement webpage for house sale. I scraped the data by using chromedriver, and renamed and dropped some of the columns. This part took high amounth time. At the end I have the data frame below. If you want to look how I scraped the data from the page you may look Scraping.ipynb.

-Resource: https://www.sreality.cz/

Here I would like to decribe type of the houses in Czech Republic. The naming could be different than other countries.

B3.Venues data in municipal parts

I used Foursquare API for getting the data.

C.Methodology

C.1.Municipal Parts of Prague

In this section, I created a foluim map to observe Prague Municipal Parts and created a gif for my blog.

Plain Prague Map

Prague Municipal Parts

You can scroll around map to observe municipal parts.

Population by Municipal Part

Praha 4 and 10 are the most crowded municipal parts of Prague.

C.2.House Sale Prices

Retrieve the House Sale Prices data.

Observe the house types.

Make latitude and longitude numeric

Check number of objects (advertisement)

Check NaN values

Correct NaN values

Add region column to the dataframe by using json file

Now we have Region column

Drop Title column

Export clean data

Exclude the municipal parts having advertisements less than 17

The houses in Praha 1 and Praha 6 are more expensive than others.

The houses in Praha 1 and Praha 6 are larger than others.

2+kk and 3+kk is most prevelant house types in Prague

Create a data frame for most expensive 20 houses in Prague

The most expensive houses are mainly placed in Praha 1.

Regression

I will look for correlation between price and other factors in this section.

C.3.Compare Municipal Parts

I will compare and cluster region in this section.

D.4.Results and Discussion

E.4.Conclusion